Still trying to Communicate with APC Smartups

[zorandinho]

Just Microsoft Composite Battery.

[Nickleback]

Hi Phillip and Fellow LabView users.

Just want to say this example works nicely with Win 7 64 bit. ( we had to un-install the APC PowerChute as mentioned the earlier post)

Huge thanks for this. you have saved a lot of hassle

Regards

Chuck

[RoyA]

Nickleback, or anyone else who can help,

Which version of the VIs did you use with Win7, 64-bit?  Were you using LabVIEW 2012? I have tried all three with no success in 2012.  I have checked everything mentioned in this thread (no APC software, never was installed,; correct battery listed; etc).  I tried running them in their original form, saving in 2012, even creating a project and copying the files out of the library.  All with no success.  Nothing changes except the 'Charging' LED (which seems to work correctly.

Here is my system:

DELL Latitude E6420 Laptop

Intel(R) Core(TM) i5-2520M CPU @ 2.50GHz   2.50 GHz

Windows 7 Enterprise 64-bit

LabVIEW Developers Suite 2012

[DavidBoyd]

RoyA,
I'm running W7x64 on a Latitude E6510. Haven't tested my code on this platform. Will try tonight or tomorrow (when I can) and get back to you.

Dave

[DavidBoyd]

RoyA,

I don't see any issues on my aforementioned Dell E6410, running W7 x64 - my code still works.  However, I'm running under the LV2011 development enironment - not 2012.  I don't have 2012 installed on this machine at present.

So it could be:

- a LV 2012 vs 2011 issue

- a rights/UAC issue (I am logged in with local admin rights)

- a problem with your .NET installation

- a problem with calling GetSystemPowerStatus from kernel32.dll on your platform

- something else I haven't thought of

Can you confirm what LV code you have and/or from where you got it?  My code is still saved in 8.2.

It registers a .NET callback to get PowerModeChanged from Microsoft.Win32.SystemEvents

The actual callback VI calls GetSystemPowerStatus from kernel32, then posts the results to a LV user event.

Let me know what else I can try to diagnose this for you.

Dave

[RoyA]

Dave,

Wow! Thanks for the quick response.  Here is where I got the three versions of code I have tried:

First, most simple - http://sine.ni.com/devzone/cda/epd/p/id/439

Second, Notifier version - https://decibel.ni.com/content/docs/DOC-1154

Third, Events version - https://decibel.ni.com/content/docs/DOC-9256

All three work, somewhat, on my DELL Latitude D830 running WXP Pro with most recent patches, and LV 2011 SP1 f2.

By somewhat I mean the tank and the High, Low, Critical, and No Battery LEDs never change.  The Power BOOLEAN and the Charging LED change properly.  The Estimated Time updates, but has 'strange' numbers (like the following sequence 336:54:37 / 00:00:00, 848:55:14, 298:36:24 / 00:00:00,  298:36:12 / 00:00:00, 296:33:04 / 00:00:00, ...)  Second number never changes.

Akso, when the UPS dies completely the Power BOOLEAN goes back to AC Power.

Thanks again.

Roy

[DavidBoyd]

My bad - I *AM* seeing issues similar to what you're reporting - I just didn't wait long enough.

I do an initial poll for current power status, which works.  And if I pull and re-plug the DC power cord on the back of my laptop, events fire, so I see the changes in AC line status and battery charging state.  But I am *NOT* getting events when the battery charge level changes.  This used to work!

So I'm puzzled.  Too tired to troubleshoot further tonight, maybe I can figure it out tomorrow.  But at least I can confirm what you've observed.

Dave

[Nickleback]

Hello Folks

I think this has moved on from the post I was going to leave as my humble offering doesn't do events but I will leave it any way as it may be of use.

Chuck

[PhillipBrooks]

Well I'm an old geezer still limping along on Windows XP and LV 8.6.1 (I do have LV2012 installed, just don't use it)

My example took the original polling example and registered for a .NET callback of PowerModeChanged. When the event occurs, the kernel32  GetSystemPowerStatus function is called to return the cluster of data containing the % values data.

Maybe you need to change the CLF Node in the PowerModeChanged to point to the correct version of kernel32 based on your OS and LabVIEW version? There are a couple of CLF nodes; one in the example, one in dotNET PowerModeChanged Handler.vi and one in dotNET PowerModeChanged.vi

[PhillipBrooks]

I just noticed that the original GetSystemPowerStatus cluster that I blindly  saved as a typdef configures the BatteryLifeTime and BatteryFullLifeTime values as I32, but the M$ descriptions for this structure indicate DWORD (unsigned).

http://msdn.microsoft.com/en-us/library/windows/desktop/aa373232(v=vs.85).aspx

Maybe the combination of the wrong data type in the cluster and 32 vs. 64 bit explains the strange values you are getting; especially if your UPS is large enough to let you run for in excess of 9 hours...

EDIT:

Help indicates the function returns -1 if time (secs) is unknown. So it must be I32. Maybe still something to do with 32/64 bit and selecting the correct instance of kernel.dll?